Add new module autopkgtest.py with the logic for determining the tests for a
source package, requesting tests via AMQP, fetching results from swift, and
keeping track of pending tests between run. This also caches the downloaded
results from swift, as re-dowloading them all is very expensive.
Integrate this into britney.py:
* In should_upgrade_src(), check whether a package is built everywhere and
installable (run_autopkgtest), and good enough to run autopkgtests for it
and its reverse dependencies.
* In write_excuses(), generate test requests for all excuses and create blocks
for those that cause test regresssions.
This introduces two new hints:
* force-badtest pkg/ver: Failing results for that package will be ignored.
This is useful to deal with broken tests that get imported from Debian or
are from under-maintained packages, or broke due to some infrastructure
changes. These are long-lived usually.
* force-skiptest pkg/ver: Test results *triggered by* that package (i. e.
reverse dependencies) will be ignored. This is mostly useful for landing
packages that trigger a huge amount of tests (glibc, perl) where some tests
are just too flaky to get them all passing, and one just wants to land it
after the remaining failures have been checked. This should be used rarely
and the hints should be removed immediately again.
Add integration tests that call britney in various scenarios on constructed
fake archives, with mocked AMQP and Swift results.
We want to treat linux-$flavor and linux-meta-$flavor as one set in britney
which goes in together or not at all. We never want to promote linux-$flavor
without the accompanying linux-meta-$flavor.
Introduce a synthetic linux* → linux-meta* dependency to enforce this grouping.
We don't use os.makedirs(dir, exist_ok=True) as that is too strict: it fails if
the directory already exists with different permissions (e. g. with 775). Thus
introduce a helper function ensuredir().
Partially revert commit ac66e311 which caused packages with unsatisfiable
dependencies to only get rejected if they were not in testing. In Ubuntu we
always want to block those.
Strip of Multi-Arch qualifiers ":any" and ":native" when building the
dependency fields, as they are not part of the package name.
This will fix cases like
Package: ipython3
Depends: python3:any (>= 3)
and include ipython3 in python3's reverse dependencies.
Closes: #794194
FAILED/SUCCESS lines would be separated by a whitespace from the list
of architectures, but not itself followed by whitespace. This is slightly
confusing, as one could interpret it as being a heading for the following
block of tested packages, rather that the final result of the previous
block.
This includes refining "HINTS_ALL" to cover all hints added at
runtime.
Currently, it is not very useful. However, a later commit will allow
policies to use this feature.
Signed-off-by: Niels Thykier <niels@thykier.net>
For items not having an age requirements (e.g. urgency=critical)
always list the "missing build" note if present.
Signed-off-by: Niels Thykier <niels@thykier.net>
This hint will block all "new" source migrations. Source migrations
for packages already in testing will be affected by this. As usual,
this hint can be overruled by an unblock hint.
Closes: GH#8
Signed-off-by: Niels Thykier <niels@thykier.net>
Reduce a "loop over all valid items" to a "loop over an item's
dependencies + reverse dependencies". For sparse graphs, this
is much more efficient.
Signed-off-by: Niels Thykier <niels@thykier.net>
Martin Pitt
Martin Pitt
Dimitri John Ledkov
Colin Watson
Robert Bruce Park
Niels Thykier
Mathieu Trudel-Lapierre
Julien Cristau